Influence of individual tree characteristics, spatial structure and logging history on tree-related microhabitat occurrence in North American hardwood forests

摘要

Background:Tree-related microhabitats(hereafter,"TreMs")are key components of forest biodiversity but they are still poorly known in North American hardwood forests.The spatial patterns of living trees bearing TreMs(hereafter,"TreM-trees")also remain to be determined.As logging practices can lead to a loss of TreM-trees and of their associated biodiversity,it is essential to identify the factors explaining TreM occurrence to better integrate them into forest management.We therefore inventoried TreMs in 40.5-ha survey strips in northern hardwood forests in Quebec,Canada,while recording the spatial location of each tree.Two strips were located in unmanaged oldgrowth forests,and 2 were in forests managed under selection cutting.All 4 stands were dominated by sugar maple(Acer saccharum Marsh.)and American beech(Fagus grandifolia Ehrn.).Beech bark disease,an exotic pathology,was observed in all the strips.Results:Large diameter at breast height and low tree vigor were the main characteristics explaining the presence of TreMs at the tree scale.TreM-trees presented slight spatial aggregation patterns.These aggregates,however,were not well-defined and were generally constituted by a large number of trees bearing few different types of TreMs.Two TreM classes(broken branch or top and woodpecker lodge)also presented a spatial aggregation.Logging practices had no significant effect on TreM occurrence.Beech bark disease increased the frequency of senescent beeches.The impact of this pathology on TreMs was however mitigated by the small size of infected trees and probably by the short time elapsed since its appearance.Conclusion:The factors explaining the presence and abundance of TreMs on trees has so far been little studied in North American hardwood forests.Our results highlight that TreM-tree characteristics in the surveyed forests are consistent with those of previous studies conducted in other forest types and regions(e.g.,Europe or Northwestern America).To our knowledge,this study is also the first to identify a spatial aggregation of TreM-trees and of specific TreM classes.It will be nevertheless necessary to determine whether the small impact of logging activities we observed results from current or past management practices.